PROJECT SUMMARY The quantitative, model driven approaches that constitute the underpinning of systems biology are emerging as an increasingly critical methodological repertoire for a truly ?precise? implementation of precision cancer medicine. This is especially relevant in view of the increasing limitations of current approaches based on the oncogene addiction paradigm. Even though pharmacological inhibition of oncogenes harboring activating alterations has emerged as a valuable rationale for targeted therapy, >75% of all adult malignancies lack any actionable alteration or present with undruggable ones and inhibitors of canonical oncogenes have shown lackluster response in the clinic. Most critically, following initial and at times remarkable response, targeted therapy almost invariably leads to relapse to drug-resistant disease. Systematic treatment of hundreds of cell lines with hundreds of compounds has shown that, with few notable exceptions, mutations are far from representing optimal predictors of targeted agent sensitivity. This is not surprising, as drug sensitivity clearly represents a complex polygenic phenotype, requiring equally complex and tumor-specific models. This center proposal encompasses studies across multiple levels of granularity, representing the full complexity of the tumor phenotype: from tumor/microenvironment interactions to single cell plasticity, supporting tumors reprograming to distinct isogenic states associated with progression or drug resistance. Specifically, three complementary directions will be pursued: First, elucidation of the regulatory module architecture (tumor checkpoint) and specific proteins within these modules (master regulators) that comprise the dysregulated mechanisms presiding over tumor homeostasis (I.e., a cell's ability to maintain its tumor state independent of mutational landscape and endogenous/exogenous signal heterogeneity). This will be accomplished by developing model-based approaches to analyze omics data representing distinct compartments of the tumor, ranging from tumor bulk, to stroma/tumor compartments, to single cells, following physiologic, genetic, and pharmacologic perturbations. Second, study the mechanisms by which tumor state can be altered to induce progression or drug resistance by adopting and extending approaches for the study of physiologic differentiation and reprograming. This analysis will integrate subclonal genomic characterization, using innovative computational models, with single cell data from primary tumors and patient derived xenografts to elucidate the mechanisms presiding over tumor plasticity. Finally, using the mechanistic regulatory frameworks emerging from these studies to elucidate actionable tumor dependencies leading to irreversible collapse of tumor homeostasis in vitro and in vivo. This will be accomplished by assembling and experimentally validating both probabilistic and kinetic models of tumor checkpoint regulation, assembled from time-series data following systematic small molecule perturbations. Center-developed software and methods will be disseminated to the Research Community, using proven strategies.